A hybrid MINres and hidden layer traversal method to improve extreme learning machine for electric load forecasting

In recent years, the Extreme Learning Machine (ELM) has emerged as a widely used neural network prediction model, demonstrating significant potential in the domain of power load forecasting. Nevertheless, conventional ELM prediction models struggle with key challenges, such as the optimal selection of hidden layer input nodes and mitigating overfitting in results. In this study, we propose a novel hybrid ELM prediction framework, integrating MINres with hidden layer traversal, to substantially enhance the accuracy of power load forecasting.